Soil depth mapping and its linkage to gully erosion rate prediction in granite areas of southern China

Abstract

Gully erosion, a spectacular erosional landform in the hilly areas of southern China, poses an enormous threat to local sustainable economic development and ecological security. The depth of granite weathering mantle increases from north to south, which is consistent with the increase of gully density and accelerated soil loss. However, the spatial distribution of soil depth in gully areas remains unclear. Herein, soil depth was measured by field sampling investigation, and soil depth mapping was established using the Fuzzy c-means algorithm. To explore the relationship between soil depth and gully erosion rate, gully characteristics and erosion process were monitored by unmanned aerial vehicle. The results showed that the gully density increased from Tongcheng (29°2′−29°24′ N) to Wuhua (23°23′−24°12′ N) in the range of 11.25–61.75 gullies/km2, in contrast to an opposite trend in the soil depth range from 180 to 5 cm. Additionally, soil depth mapping by fuzzy c-means algorithm performed well in gully areas, with overall accuracy, Pearson correlation coefficient, and agreement coefficient reaching 81.60%, 0.90, and 0.97, respectively. Moreover, the gully erosion rate in the range of 7.60–29.08 m3 yr.−1 had a significant positive linear relationship with the area of collapsing wall (R2 = 0.54, p < 0.05). Based on this relationship, a power function was established between soil depth and gully erosion rate (R2 = 0.41, p < 0.01), which provides a possibility for effective prediction of gully erosion. This study facilitates a better understanding of the gully erosion mechanism and its effective control.